There are no longitudinal data that address weight loss stability and lipid levels in bariatric surgical patients. The goal of this study was to determine whether weight regain adversely affected reduction in lipid levels after gastric bariatric operations. Of 651 consecutive patients undergoing gastric restrictive surgery for morbid obesity, 227 (35%) had increased serum levels of total cholesterol (TC), triglycerides, or both preoperatively. High-density lipoprotein cholesterol (HDL-C) levels were subnormal (</=35 mg/dl) in 45 (20%) of the hyperlipidemic patients. Fasting lipid profiles were determined at 6-month intervals postoperatively. This series included the following three operations: gastroplasty (GP; N = 13), standard Roux-en-Y gastric bypass (RYGB; N = 205), and distal Roux-en-Y gastric bypass (DRY; N = 9). By 6 months postoperatively, patients had a >/=15% mean reduction in TC and a >/=50% mean reduction in triglycerides, both of which were significant in comparison with preoperative levels (P </=0.05). Mean HDL-C levels had increased significantly vs. preoperative levels by 12 months postoperative y (P <0.05) and continued to increase through 5 years. By 18 months both HDL-C and TC were significantly lower after DRY than after GP or RYGB. In 91 patients who were followed for 2 years or longer (mean 48 +/- 25 months), mean excess weight loss was 55% with mean body mass index reduced from 48 to 33 kg/m(2). This group was divided into patients whose weight remained stable (N = 54) and patients who regained >/=15% of their lost weight or lost less than 50% of excess weight (N = 37). Although mean excess weight loss and body mass index were significantly different between the two groups (P <0.0001) at 2 years, there was no difference in the lipid profile (TC/HDL) between the two groups at any interval through 5 years. These results show that abnormal lipid profiles can be permanently improved after gastric bariatric surgery and are not adversely affected by mediocre weight loss or regaining >/=15% of lost weight. DRY appears to be a superior operation for TC reduction in comparison with GP and RYGB.